CV/AI Software Engineer
- Posted
- Employment type
- Full-time
Who we’re looking for
- We’re looking for a Prague-based Computer Vision / AI Engineer to build and optimize real-time traffic safety algorithms that run on edge devices with limited compute.
- You’ll adapt and fine-tune modern neural network architectures, optimize them for performance (including quantization), and ship robust software that handles challenging real-world video conditions—even corrupted inputs—on specialized hardware accelerators.
What you’ll do
- Develop and deploy real-time computer vision algorithms for traffic safety in real time.
- Optimize end-to-end performance to maximize hardware utilization with limited compute capacity.
- Customize minimal system images for edge devices.
- Adapt, fine-tune, and evaluate modern neural network architectures for our production use cases.
- Apply model optimization techniques (including quantization) to achieve the best accuracy-to-performance ratio.
- Developing tools to read even corrupted videos.
- Dealing with interesting hardware quirks of hardware accelerators.
- Collaborate with engineers and stakeholders to define requirements, measure results, and iterate.
About the position
At GoodVision, a trusted provider in AI-driven traffic and intelligent transport solutions, we are building an advanced data analytics platform that transforms real-world movement data into digital insights. Our platform predicts future trends, empowering decision-makers with actionable insights to improve traffic management and safety. Our mission aligns with the VISION ZERO strategy, dedicated to eliminating all traffic fatalities and severe injuries while enhancing safe, healthy, and equitable mobility for all. Through AI, we are reshaping urban dynamics towards more livable cities by connecting data, processes, and expertise. GoodVision’s solutions are trusted globally, with >100 enterprise clients across >50 countries leveraging our platform to monitor, digitize, analyze, and act to reduce congestion, prevent incidents, and improve commute times worldwide.